Originally published Sep 2024. Updated June 2026.
Booking.com is the most-visited travel website in the world. Over 400 million monthly visits. Parent Booking Holdings reported $23.7 billion in 2024 revenue, up 11 percent year over year, and a Q1 2026 net income margin of 19.6 percent.
It did not get there on luck. It got there on a specific, sustained, decade-plus marketing operating system that is now being rebuilt around AI.
From 2010 through 2022, Booking.com ran the most disciplined performance-marketing machine in consumer commerce. Five components.
1. Bid-on-everything Google Ads. Booking was one of Google's largest advertisers globally for over a decade. The company bid on millions of long-tail travel queries — every city, every hotel chain, every property type, every date combination. The volume gave Booking better data on conversion rates than almost anyone else in commerce, which in turn made its bids more efficient than anyone else's.
2. Aggressive retargeting. If you looked at a hotel and didn't book, Booking found you again — on Facebook, Instagram, display networks, and email — usually with that exact hotel, often with a price drop or urgency message. End-to-end automated, tuned by full-time data science teams.
3. Relentless A/B testing. Booking ran tens of thousands of concurrent A/B tests on its own site at any moment. The cultural commitment to experimentation produced UX improvements competitors couldn't replicate without the same testing infrastructure. The conversion advantage was measurable and durable.
4. Localization at industrial scale. Booking now supports over 100 payment methods, 50+ currencies, and operates in 220+ countries and territories. The localization moat is invisible to most marketers but real — Booking captured cross-border traffic competitors could not convert.
5. Personalized email and dynamic content. Behavioral email — destinations the user searched, properties they viewed, dates that aligned with prior trips — drove repeat-booking rates significantly above the OTA average. The email program alone is responsible for billions in annual booking volume.
What's changing in 2026
The classical OS is being restructured around AI for two reasons. The discovery layer Booking depended on — Google search — is being absorbed by AI answer products that bypass the OTA referral. And AI agents are becoming the new interface for planning and booking, which means Booking has to be the agent or be inside the agent that wins.
Booking's response, articulated in its 2026 shareholder letter, is the "Connected Trip vision." The strategic intent is to use Generative AI to deliver an integrated trip-planning, booking, and management experience that covers flights, hotels, rentals, experiences, and OpenTable reservations as a single connected product.
The execution involves three layers. An AI agent inside Booking.com that handles natural-language planning queries end to end. Faster customer service via GenAI-augmented response systems — the 2026 letter explicitly cited "accelerating the integration of Generative AI to improve customer service response times." And improved partner tools that use AI to give supply-side partners better pricing, content, and conversion guidance.
What competitors are doing differently
Expedia is running a similar AI-agent strategy with a different architecture — pushing harder into Vrbo and short-term rentals, integrating AI into the Hotels.com flow. The two companies are converging on similar product directions from different starting positions.
Airbnb is doing something different — using AI to expand its category footprint (Experiences, Services) rather than to optimize a fixed-category booking flow. The product divergence reflects the strategic divergence.
Google is the structural threat — building integrated travel answers and bookings inside AI Overviews and Gemini that absorb the discovery layer Booking has historically depended on for paid traffic.
Lessons for other travel marketers
Three things to copy.
The testing culture. Booking's A/B testing infrastructure took a decade to build and now produces conversion improvements competitors can't match. Most travel brands run a fraction as many tests as they should.
The localization investment. Cross-border travel is most travel. Brands that have not invested in real localization — language, payment methods, currencies, local partnerships — leave significant revenue uncollected.
The AI agent commitment. The brands that ship a credible AI agent inside their own customer experience in 2026 will own the relationship in the AI era. The brands that wait will service customers who arrived through someone else's agent.
FAQ
Q: What makes Booking.com's marketing different from competitors?
Five things: industrial-scale Google Ads bidding, aggressive retargeting, relentless A/B testing, deep localization, and personalized email at scale. The components are individually replicable. The compounding effect from running all five together for a decade is not.
Q: How much revenue does Booking Holdings generate?
$23.7 billion in 2024, up 11 percent year over year. Q1 2026 net income margin: 19.6 percent. The company executed a 25-for-1 stock split on April 2, 2026 and has $18.2 billion in remaining buyback authorization.
Q: What is the Connected Trip vision?
Booking's strategic plan to use Generative AI to integrate flights, hotels, rentals, experiences, and OpenTable into a single connected trip-planning and booking experience.
Q: How is Booking responding to AI Overviews and ChatGPT?
By building its own AI agent inside Booking.com and integrating GenAI across the customer experience. The logic is to become the agent — or to be inside whichever agent wins — rather than depending on referral traffic from search engines AI is now absorbing.
Q: What can smaller travel brands learn from Booking?
Build a real A/B testing culture, invest in genuine localization, and ship a credible AI agent inside your customer experience in 2026.
Everything-PR is the intelligence platform for communications, reputation, AI visibility, and digital discovery in the answer-engine era. Thirty-plus publications. Publishing since 2009. Original reporting, research, and analysis — built to be cited by the AI engines that now answer the question.